Purnamasari, Ni Made Gita Dwi (2018) Identifikasi Tweet Cyberbullying pada Aplikasi Twitter menggunakan Metode Support Vector Machine (SVM) dan Information Gain (IG) sebagai Seleksi Fitur. Sarjana thesis, Universitas Brawijaya.
Abstract
Cyberbullying merupakan salah satu tindakan yang melanggar UU ITE dimana kejahatan ini dilakukan di media sosial salah satunya aplikasi Twitter. Tindakan ini sulit terdeteksi jika tidak ada yang me-report tweet tersebut. Identifikasi tweet cyberbullying bertujuan untuk mengklasifikasikan tweet yang mengandung konten bullying pada akun Bapak Fadli Zon selaku Wakil Ketua DPR RI. Klasifikasi dilakukan dengan menggunakan metode Support Vector Machine (SVM) dimana metode ini akan mencari hyperplane pemisah antara kelas negatif dan positif. Penelitian ini merupakan klasifikasi teks dimana semakin banyak datanya semakin banyak fitur yang dihasilkan, oleh karena itu penelitian ini juga menggunakan seleksi fitur Information Gain (IG) untuk menyeleksi fitur yang tidak relevan terhadap klasifikasi. Dimana nilai information gain akan diurutkan secara ascending untuk fitur diseleksi berdasarkan nilai threshold yang telah ditentukan. Nilai information gain akan bernilai tinggi jika term merepresentasikan satu kelas dengan hanya muncul sekali atau lebih pada kelas tersebut. Proses sistem dimulai dari text preprocessing dengan tahapan tokenisasi, filtering, stemming dan pembobotan kata. Kemudian melakukan seleksi fitur information gain dengan menghitung nilai entropy tiap kata. Setelah itu melakukan proses klasifikasi berdasarkan fitur yang telah diseleksi dan hasil keluaran sistem berupa identifikasi apakah tweet termasuk bully atau bukan bully. Hasil akurasi menggunakan metode SVM dengan accuracy 75%, precision 70.27%, recall 86.66% dan f-measure 77.61% pada percobaan nilai iterMax = 20, λ = 0.5,
English Abstract
Cyberbullying is one of the actions that violate the ITE Law where the crime is committed on social media applications such as Twitter. This action is difficult to detect if no one is reporting the tweet. Cyberbullying tweet identification aims to classify tweets containing bullying on Mr Fadli Zon account who is the Vice Chairmaman of the Indonesian Parliament. Classification is done using Support Vector Machine method where this method aims to find the dividing hyperplane between negative and positive class. This study is a text classification where more data is used, the more features are produced, therefore this research also uses Information Gain as feature selection to select features that are not relevant to the classification. The entropy of feature will be sorted ascending for the selected feature based on the predefined threshold value. The entropy will be of high value if the term represents a class by appearing only ne or more in that class. The process of the system starts from text preprocessing with tokenizing step, filtering, stemming and term weighting. Then perform the information gain feature selection by calculating the entropy value of each term. After that perform the classification process based on the features that have been selected, and the output of the system in the form of identification whether the tweet is bullying or not. The result of accuracy using SVM method with accuracy 75%, precision 70.27%, recall 86.66% and f-measure 77.61% on experiment iterMax value = 20, λ = 0.5, γ = 0.001, ε = 0.000001, and C = 1. And threshold value the best information gain feature selection is 90%, with accuracy value 76.66%, precision 72.22%, recall 86.66% and f-measure 78.78%
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | SKR/FTIK/2018/405/051806608 |
| Uncontrolled Keywords: | Cyberbullying, Klasifikasi, Support Vector Machine, Information Gain |
| Subjects: | 300 Social sciences > 302 Social interaction > 302.3 Social interaction within groups > 302.34 Social interaction in primary groups > 302.343 Bullying |
| Divisions: | Fakultas Ilmu Komputer > Teknik Informatika |
| Depositing User: | Budi Wahyono Wahyono |
| Date Deposited: | 29 Nov 2018 07:04 |
| Last Modified: | 18 Oct 2021 03:40 |
| URI: | http://repository.ub.ac.id/id/eprint/161608 |
Preview |
Text
Ni Made Gita Dwi Purnamasari.pdf - Published Version Download (24MB) | Preview |
Actions (login required)
 |
View Item |